Candidate Goals and Career Development Plan: The candidate is an interventional neuroradiologist with a long term goal of becoming an independent investigator with clinical responsibilities in an academic medical center. The candidate has learned basic skills in the use of Positron Emission Tomography (PET) for cerebral hemodynamic and metabolic measurements. This award will secure the dedicated time and guidance needed to develop the foundation of knowledge and skills required for successful long term investigation. This foundation will be built over the proposed three years of the award by (1) gaining a deeper understanding of cerebrovascular pathophysiology and the use of PET for it's study through participation in a NIH funded Program Project; (2)learning advanced skills in physiologic magnetic resonance (MR) measurements and (3) gathering preliminary data for the development of an independent project Environment: The environment is unique in the availability of both PET and MR expertise and technology. The candidate's sponsor is well recognized as an independent investigator of cerebrovascular pathophysiology using PET. Similarly, the co-sponsor is director of an active MR research group, recognized for work in physiologic MR applications. A funded Program Project using PET to study cerebral hemodynamics and metabolism in patients with acute neurologic injury will be underway during the three year term of this award. Research Proposal: The primary aim is to investigate the use of MR for the measurement of regional cerebral oxygen extraction fraction (rOEF) in humans. Preliminary data indicate that these measurements have important clinical significance. The primary hypothesis to be tested is that deoxyhemoglobin sensitive MR maps (known as blood oxygen level dependent, BOLD or R2* maps), after correction for blood volume, will identify regional increases in rOEF. First, the accuracy of MR measurements of cerebral blood volume (rCBV) will be assessed. This will be accomplished with paired MR and PET studies in 30 patients with carotid occlusion. Next, the relationship of rOEF measured by direct jugular A-VDO2 sampling and rCBV corrected blood oxygen level dependent MR maps will be rigorously examined in a series of Macaque monkey experiments. Finally, the hypothesis that a blood oxygen level dependent map corrected for rCBV will accurately identify regional changes in rOEF will be tested in 30 patients with stable carotid occlusion studied with PET and MR.